A stereo vision-based obstacle detection system in vehicles

Huh, Kunsoo; Jaehak, Park; Hwang, Jihyun; Hong, Daegun

doi:10.1016/j.optlaseng.2007.08.002

Cited by 57 publications

(26 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The final matching cost, MCfinal, which indicates dissimilarity between the left and right pixels, is calculated using the support weights in the left and right support windows and the calculated RMCs. It is expressed as: (2). After the final matching costs related to all possible disparities are calculated, the final disparity dfinal is determined by the winner-takes-all (WTA) method [4] and is expressed as:…”

Section: Adaptive Census Transformmentioning

confidence: 99%

“…To obtain 3D information, it calculates the disparities between the matching points of a stereo image pair. The obtained 3D information can be used in a number of applications, such as automotive vehicles, robot vision, monitoring systems, mobile applications, and 3D reconstruction [1][2][3]. Crucial to the reliability of 3D information in these applications is the accurate calculation of disparity that is carried out by two types of stereo matching algorithms: local and global matching algorithms [4,5].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Improved Stereo Matching Algorithm with Robustness to Noise Based on Adaptive Support Weight

2017

J Inf Process Syst

View full text Add to dashboard Cite

An active research area in computer vision, stereo matching is aimed at obtaining three-dimensional (3D) information from a stereo image pair captured by a stereo camera. To extract accurate 3D information, a number of studies have examined stereo matching algorithms that employ adaptive support weight. Among them, the adaptive census transform (ACT) algorithm has yielded a relatively strong matching capability. The drawbacks of the ACT, however, are that it produces low matching accuracy at the border of an object and is vulnerable to noise. To mitigate these drawbacks, this paper proposes and analyzes the features of an improved stereo matching algorithm that not only enhances matching accuracy but also is also robust to noise. The proposed algorithm, based on the ACT, adopts the truncated absolute difference and the multiple sparse windows method. The experimental results show that compared to the ACT, the proposed algorithm reduces the average error rate of depth maps on Middlebury dataset images by as much as 2% and that is has a strong robustness to noise.

show abstract

Section: Adaptive Census Transformmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

An Improved Stereo Matching Algorithm with Robustness to Noise Based on Adaptive Support Weight

2017

J Inf Process Syst

View full text Add to dashboard Cite

show abstract

“…The software mainly consists of three parts, such as graying module, median filter module and binarization module [5,6]. Then the processor will detect lanes and the position of the vehicle in the image is calculated.…”

Section: System Overviewmentioning

confidence: 99%

Design of Lane Departure Warning System Based on ADSP-BF561

Liu¹,

Lang²,

Qiao³

2012

Proceedings of the 2nd International Conference on Electronic and Mechanical Engineering and Information Technology (2012)

View full text Add to dashboard Cite

Abstract. Aiming at the problem of more and more traffic accidents which are caused by the driver behaviors, this paper designs a lane departure warning system based on the ADSP-BF561 processor. Vision sensor is used to capture road images, while the driving parameters are collected by the microprocessor. The lane marking detection speed can be more than 25 frames per second. The system is equipped with the test vehicle for hundreds of kilometers on highway, and the results show that this system has a better efficiency and feasibility. IntroductionWith a large increase of the vehicles all over the world, the traffic transportation safety problems have attracted more and more people's extensive attention. According to statistics, 80% of the traffic accidents were caused by the driver's subjective factors, such as driver distraction, negligence, and fatigue [1]. If the warning is given one second ahead of danger, 90% of traffic accidents can be avoided [2]. Thus, reducing traffic accidents caused by the driver factors is the focus of the transportation security. On one hand, drivers need to strengthen their literacy of traffic rules; on the other hand, technical measures should be taken on the driver's dangerous behavior for early warning. The lane departure warning system installed on the vehicle is a very effective way [3].The traditional lane departure warning systems are based on IPC platforms, which are not only bulky and costly, but also not suitable for marking. In this paper, we design a lane departure warning system based on the vision sensor and ADSP-BF561 processor, which has a small, high processing speed. The system will not only be able to solve some illegal behaviors, such as driving across the lane, but also can provide detailed information for the traffic management department to restore the scene of the accident [4].

show abstract

“…Huh ve ark. [4] stereo görme sensörleri kullanarak bir engel tanıma sistemi geliştirmiştir. Bu sistemde başlangıçta eşleşen çiftlerin sağlıklı bir şekilde tespit edilebilmesi için özellik eşleştirme, epipolar sınır ve özelliklerin toplanması işlemleri kullanılmıştır.…”

Section: Gi̇ri̇ş (Introduction)unclassified

Stereo kamera sisteminde aykırılık haritaları yardımıyla nesne uzaklıklarının tespit edilmesi

Soyaslan

2015

SAUFenBilDer

View full text Add to dashboard Cite

ÖZBu çalışmada stereo kamera sistemi ile farklı cisimlerin derinliklerinin yani paralel kamera düzlemine uzaklıklarının tespit edilmesi amaçlanmıştır. Kurulan test düzeneğinde sağ ve sol görüntünün elde edilmesi için iki adet sabitlenmiş kamera ve bunlardan alınan görüntüleri işleyen bir adet bilgisayar bulunmaktadır. Kameralar sabit olarak yerleştirildikten sonra satranç tahtası yardımıyla kalibre edilmiştir. Kalibre işleminden sonra her iki kamera için esas ve temel matrisler, birinci ve ikinci kameralar arasında dönme ve öteleme matrisi bulunmuştur. Ardından sol ve sağ kameradan alınan iki adet görüntü üzerinde doğrultma işlemi gerçekleştirilmiştir. Buna bağlı olarak derinlik haritası ve uzaklık bilgisine ulaşılmıştır. Derinlik bilgisi bulma işleminin robot ve araçlardaki uzaklık algılama sistemlerinden engelliler için uzaklık tayinine kadar birçok alanda kullanılması mümkündür.Anahtar Kelimeler: stereo görme, derinlik haritası, görüntü işleme, uzaklık tayini Object distance detection through disparity map in stereo camera system ABSTRACT In this study, it is intended to detect various objects' depths consequently distances from parallel camera plane with stereo camera system. In the established test setup, there are two fixed cameras for capturing left and right image and a computer for processing the images. After cameras were placed fixedly, they were calibrated with chessboard. Subsequently calibration; the intrinsic parameters, essential and fundamental matrices, rotation and translation matrices between left and right cameras were occurred. Then rectification process performed to the images which captured from left and right cameras. Accordingly, depth map and knowledge of distance has been obtained. It is possible to use the depth knowledge process in many fields involved robot and vehicle distance detect systems to handicapped distance determination.

show abstract

A stereo vision-based obstacle detection system in vehicles

Cited by 57 publications

References 10 publications

An Improved Stereo Matching Algorithm with Robustness to Noise Based on Adaptive Support Weight

An Improved Stereo Matching Algorithm with Robustness to Noise Based on Adaptive Support Weight

Design of Lane Departure Warning System Based on ADSP-BF561

Stereo kamera sisteminde aykırılık haritaları yardımıyla nesne uzaklıklarının tespit edilmesi

Contact Info

Product

Resources

About